Nonlinear System Identification Using Robust Fusion Kernel-Based Radial basis function Neural Network

Abstract

In this paper, authors have proposed, a robust Fusion kernel-based Radial basis function (RBF) neural network algorithm is proposed. The objective is to solve dynamic nonlinear system identification problems. The proposed algorithm is a Fusion of both the Gaussian and Cosine, which is capable of updating the weight of kernels by using the gradient descent method. The weight updating process enhances its adaptive learning capabilities. The proposed model is further tested with an ARMA model to prove its superiority. The comparison results illustrate, the proposed method achieves good performance over other models. The performance is evaluated through mean square error (MSE). IN the testing phase the model achieves a very low Mean squared error compared to the existing approach.